| --- |
| title: README |
| emoji: 🎨 |
| colorFrom: pink |
| colorTo: indigo |
| sdk: static |
| pinned: false |
| --- |
| |
| <img src="./assets/animetimm-banner.png" alt="AnimeTimm banner" width="100%"> |
|
|
| # AnimeTimm |
|
|
| **AnimeTimm is a DeepGHS project for training, testing, and sharing `timm`-based vision models for anime-style and illustration-focused image tagging.** |
|
|
| It is part research playground, part anime-fan workshop: we care about reproducible datasets, model cards, ONNX exports, and practical demos, but the models are also built for people who actually work with 2D art, tags, characters, and visual search. |
|
|
| ## Project Stewardship |
|
|
| AnimeTimm is produced and maintained by the [DeepGHS](https://huggingface.co/deepghs) team and contributors. |
|
|
| This Hugging Face organization is the focused publishing home for AnimeTimm releases: model checkpoints, selected training datasets, and interactive Spaces. The upstream engineering work is connected to the [DeepGHS GitHub organization](https://github.com/deepghs), including the [`deepghs/animetimm`](https://github.com/deepghs/animetimm) repository. |
|
|
| ## What We Build |
|
|
| - `timm`-based image tagging and classification models for anime-style images. |
| - Training datasets prepared for large-scale tagger experiments. |
| - PyTorch, Safetensors, and ONNX artifacts where available. |
| - Playgrounds and ranklists for trying models and comparing outputs. |
|
|
| ## Featured Dataset |
|
|
| ### [`danbooru-wdtagger-v4-w640-ws-full`](https://huggingface.co/datasets/animetimm/danbooru-wdtagger-v4-w640-ws-full) |
|
|
| The main public dataset release used by the `dbv4-full` model family. It is a Danbooru-derived WebDataset build for large-scale anime-style multi-label tagging, with images resized so `min(width, height) <= 640`. |
|
|
| | Split | Images | Total Size | |
| | --- | ---: | ---: | |
| | train | 5,321,713 | 318 GB | |
| | test | 295,926 | 17.7 GB | |
| | val | 296,957 | 17.8 GB | |
| | **total** | **5,914,596** | **353.5 GB** | |
|
|
| Each sample contains the image as `webp` plus JSON metadata: `id`, `width`, `height`, `rating`, `general_tags`, and `character_tags`. The selected label space has **12,476 tags**: 9,225 general tags, 3,247 character tags, and 4 rating tags. |
|
|
| ## Model Zoo: `dbv4-full` |
|
|
| The tables below focus only on the main `dbv4-full` model line. Metrics are copied from the corresponding model cards and use the test split reported there. |
|
|
| <img src="./assets/dbv4-full-model-snapshot.png" alt="dbv4-full model performance and parameter snapshot" width="100%"> |
|
|
| ### Top 5 By Macro F1 |
|
|
| | Rank | Model | Family | Params | Macro@Best F1 | Macro@0.40 F1 | Micro@0.40 F1 | |
| | --- | --- | --- | --- | --- | --- | --- | |
| | 1 | [convnextv2_huge.dbv4-full](https://huggingface.co/animetimm/convnextv2_huge.dbv4-full) | ConvNeXt | 692.6M | 0.611 | 0.580 | 0.697 | |
| | 2 | [eva02_large_patch14_448.dbv4-full](https://huggingface.co/animetimm/eva02_large_patch14_448.dbv4-full) | EVA | 316.8M | 0.599 | 0.569 | 0.693 | |
| | 3 | [caformer_b36.dbv4-full](https://huggingface.co/animetimm/caformer_b36.dbv4-full) | CAFormer | 134.0M | 0.581 | 0.546 | 0.689 | |
| | 4 | [swinv2_base_window8_256.dbv4-full](https://huggingface.co/animetimm/swinv2_base_window8_256.dbv4-full) | SwinV2 | 99.7M | 0.575 | 0.541 | 0.683 | |
| | 5 | [caformer_m36.dbv4-full](https://huggingface.co/animetimm/caformer_m36.dbv4-full) | CAFormer | 82.7M | 0.559 | 0.515 | 0.676 | |
|
|
| ### Representative Models By Backbone Family |
|
|
| Each row is the best `dbv4-full` model currently published for that backbone family. |
|
|
| | Family | Model | Params | Macro@Best F1 | Macro@0.40 F1 | Micro@0.40 F1 | |
| | --- | --- | --- | --- | --- | --- | |
| | ConvNeXt | [convnextv2_huge.dbv4-full](https://huggingface.co/animetimm/convnextv2_huge.dbv4-full) | 692.6M | 0.611 | 0.580 | 0.697 | |
| | EVA | [eva02_large_patch14_448.dbv4-full](https://huggingface.co/animetimm/eva02_large_patch14_448.dbv4-full) | 316.8M | 0.599 | 0.569 | 0.693 | |
| | CAFormer | [caformer_b36.dbv4-full](https://huggingface.co/animetimm/caformer_b36.dbv4-full) | 134.0M | 0.581 | 0.546 | 0.689 | |
| | SwinV2 | [swinv2_base_window8_256.dbv4-full](https://huggingface.co/animetimm/swinv2_base_window8_256.dbv4-full) | 99.7M | 0.575 | 0.541 | 0.683 | |
| | ViT | [vit_base_patch16_224.dbv4-full](https://huggingface.co/animetimm/vit_base_patch16_224.dbv4-full) | 95.8M | 0.540 | 0.500 | 0.664 | |
| | MobileNetV4 | [mobilenetv4_conv_aa_large.dbv4-full](https://huggingface.co/animetimm/mobilenetv4_conv_aa_large.dbv4-full) | 47.3M | 0.511 | 0.458 | 0.641 | |
| | MobileNetV3 | [mobilenetv3_large_150d.dbv4-full](https://huggingface.co/animetimm/mobilenetv3_large_150d.dbv4-full) | 29.3M | 0.462 | 0.400 | 0.605 | |
| | MobileViT | [mobilevitv2_200.dbv4-full](https://huggingface.co/animetimm/mobilevitv2_200.dbv4-full) | 30.2M | 0.454 | 0.401 | 0.608 | |
| | ResNet | [resnet152.dbv4-full](https://huggingface.co/animetimm/resnet152.dbv4-full) | 83.7M | 0.486 | 0.448 | 0.624 | |
|
|
| ## Try It |
|
|
| - [`dbv4-full-playground`](https://huggingface.co/spaces/animetimm/dbv4-full-playground) - tag images with pretrained `dbv4-full` models. |
| - [`dbv4-full-ranklist`](https://huggingface.co/spaces/animetimm/dbv4-full-ranklist) - compare the public `dbv4-full` model lineup. |
|
|
| ## Maintenance |
|
|
| The source data and chart builder are stored in this Space repository so the organization card can be regenerated without guessing: |
|
|
| - [`data/dbv4_full_models.csv`](./data/dbv4_full_models.csv) - checked-in metric table. |
| - [`data/dbv4_full_dataset_summary.json`](./data/dbv4_full_dataset_summary.json) - checked-in featured dataset summary. |
| - [`data/featured_models.json`](./data/featured_models.json) - top-5 and best-by-family selections. |
| - [`scripts/build_org_card.py`](./scripts/build_org_card.py) - regenerates the banner and model snapshot chart from the checked-in data. |
|
|
| ## Acknowledgements |
|
|
| - [@narugo1992](https://huggingface.co/narugo1992) ([GitHub](https://github.com/narugo1992)) completed the [`deepghs/animetimm`](https://github.com/deepghs/animetimm) GitHub project, built the end-to-end data, training, and release pipeline, and carried out the full model training work for AnimeTimm. |
| - [@SmilingWolf](https://huggingface.co/SmilingWolf) ([GitHub](https://github.com/SmilingWolf)) is gratefully acknowledged for the AnimeTagger idea and for providing mature Danbooru metadata cleaning techniques that made this line of work much more practical. |
| - [@7eu7d7](https://huggingface.co/7eu7d7) ([RainbowNeko / IrisRainbowNeko on GitHub](https://github.com/IrisRainbowNeko)) is gratefully acknowledged for directional guidance during training and tuning. |
| - Thanks to all other DeepGHS contributors who helped shape the surrounding infrastructure, experiments, reviews, and release process. |
|
|
| ## Notes |
|
|
| These releases are research and hobbyist infrastructure for visual tagging. Please check each model or dataset card for license, source data notes, intended use, and audience restrictions before reuse. |
|
|
